TAK Data Collection

In modern warfare, data is a weapon system. From computation to combat, the never-ending flow of information that supports the functioning of our economy, military technology, and command and control is now a domain that will increasingly determine the outcome of current and future conflict. Information dominance is the key to staying in front of our adversaries in the competitive landscape that, if managed improperly, could jeopardize our national security. Global military operations demand a strategic information dominance platform that can collect operational data to feed dynamic decision agents to ensure fight-tonight readiness in defense of our nation. These operations demand technologies that respond quickly to dynamic operations during contingency, humanitarian, and peacetime operations. Transforming data into a weapon system enables readiness, informs decision-makers, enables dominance in the information and cyberspace domains, and expands power projection Multi-Domain Operations (MDO). Analysts are provided with previously gathered intelligence from a patrol, initially from various JADC2 platforms. Still, they are often not given full details until units return to their home base/staging area. While this state does provide some intel value, it doesnĺt offer enough data to enable the data weapon system concept, maximize domain awareness, or facilitate rapid tactical decision-making. To help achieve our goals, what if we could take every edge ATAKáasset and turn them into a connected, real-time, ML-enabled data collection sensor/analyst? Data capture alone is insufficient to satiate the military desire for information dominance. The militaryĺs most important objective of data capture should be to ensure that the data gathered is reliable and packed with actionable insights that can be analyzed and turned into data-driven decisions. To realize this goal, the military must be able to create clear definitions, semantic linkages, and proper organization to harness data for mission-critical and operations-critical purposes. A fundamental transformation is needed:áa framework that seamlessly combines data definitions, collection strategies, labeling, and organizational protocols, all within the purview of a consistent ML-focused knowledge graphing service. Many options exist today for IoT data collection, including proprietary software solutions that integrate with proprietary hardware. The TPCámay not achieve an enduring solution by relying on commercial offerings due to these solutions: Being sourced from insecure supply chains Being developed by non-US partners Housing data in non-DoD compliance data centers Asserting improper rights on DoD data We propose to build the TAK ML Data Platform (TAK-MLD) with proven Government-off-the-Shelf (GOTS) and open-source tools. This enables the AFRLáand the TPCáto have maximum flexibility to mature and transition the TAK-MLD to the warfighter.á